TODO-_unsorted.md

-This should get split into other files.
 
-```
-- Improve options.
+- [ ] Defer the avformat_find_stream_info, and get basic stream info before setting options. An alternative is to provide a `av.probe` function or something.
 
-  - Per-stream options (which require not opening the codecs right away).
+- [ ] Container.decode_one(**demux_kwargs) -> Frame. There would be a Container._buffered_decoded_frames or something that would contain extras, just in case. Stream and CodecContext have the same.
 
-  - Current input behaviour:
-    - av.open(..., options={}, container_options={}, stream_options=[{}, ...])
-    - Those options are provided to each stream via avformat_find_stream_info,
-      which is called immediately. However, it isn't garunteed to open all
-      codecs, so options may not have been provided at this time.
+- [ ] Data stream support.
 
-  - Ideal input behaviour:
-    - You can defer the avformat_find_stream_info, and get basic stream info
-      before setting options.
-    - stream.codec_context.options then gets during avformat_find_stream_info
-      or open or whatever.
+- [ ] assert nb_channels vs channel_layout
 
-- AVFormatContext::interrupt_callback
-
-- Container.decode_one(**demux_kwargs) -> Frame
-  There would be a Container._buffered_decoded_frames or something that would
-  contain extras, just in case.
-
-  - Stream.decode_one() would do the same.
-
-- Data stream support.
-
-- Stream.encode(packet) should also mux frames.
-  For b/c, it should return packets with a _muxed_by_encode flag set, which then
-  has them get ignored once; the flag is reset when muxed.
-
-- assert reasonable time_base and PTS handling
-
-  AudioFifo
-
-    What does it need?
-      A: - sample_fmt
-         - channel_layout
-
-    Where does it get them?
-      The first frame it processes.
-
-
-
-  AudioResampler
-    - what does it need?
-      - in/out sample_fmt
-      - in/out channel_layout
-      - in/out sample_rate
-
-- assert nb_channels vs channel_layout
-
-- Make everything mutable. Later, we can introduce a PseudoMutable base class
-  for everything that has a .assert_mutable() method and public is_read_only
-
-- Document/test the entire API (in the most casual way).
-
-- Include build discovery info into the module, retrievable by `python -m av --version`
-
-- VideoFrame.load(file_or_path); opens the file and decodes the first frame
-
-- InternalObject class for stopping construction:
-
-    __cinit__(self, sentinel):
-        assert sentinel
-    
-    @classmethod
-    cdef alloc(cls, ...):
-        self = cls(sentinel)
-        self._init(*args, **kwargs)
-    
-- isdeferredinstance loads the module via sys.modules, then getattr on it, and
-  finally do an isinstance.
+- [ ] Frame.save(file_or_path)
+- [ ] Frame.load(file_or_path) (classmethod); opens the file and decodes the first frame. Perhaps `av.load_frame` or something.
   
-  isdeferredinstance(x, ('PIL.Image', 'Image'))
+- [ ] `isdeferredinstance` loads the module via sys.modules, then getattr on it, and
+  finally do an isinstance. `isdeferredinstance(x, ('PIL.Image', 'Image'))`
 
-- Rename Buffer and ByteSource to represent they are opposite directions of
+- [ ] Rename `Buffer` and `ByteSource` to represent they are opposite directions of
   the same operation.
 
-- `atexit.register` something to clean up FFmpeg threads
+- [ ] `atexit.register` something to clean up FFmpeg threads
 
-- Directly use the Numpy C-API
+- [ ] Directly use the Numpy C-API
   - see: http://docs.scipy.org/doc/numpy/reference/c-api.array.html
   - see: https://github.com/cython/cython/wiki/tutorials-numpy#c-api-initalization
 
-- Move the logic regard initializating a VideoReformatter into the object
+- [ ] Move the logic regard initializating a `VideoReformatter` into the object
   itself instead of the VideoFrame.reformat method.
 
-- Protocol for streams to pull packets from iterators. Then secondary streams
+- [ ] Protocol for streams to pull packets from iterators. Then secondary streams
   get easier: container.add_stream('audio_codec', source=av.open('music.aiff').iter_packets(only_primary=True))
 
-- Make av.utils.SmartPointer to wrap around a pointer? All this would do is
+- [ ] Make av.utils.SmartPointer to wrap around a pointer? All this would do is
   hold a reference to said pointer and make sure it isn't garbage collected.
 
-- Look at how stream alloc works; can we manually do this and register the
+- [ ] Look at how stream alloc works; can we manually do this and register the
   stream on a format context later?
 
-- Various AudioLayout/AudioFormat/VideoFormat attributes should be writable.
+- [ ] Various AudioLayout/AudioFormat/VideoFormat attributes should be writable.
     - is_mutable flags on various objects (including formats, layouts, contexts,
       streams, etc.) could guard the __set__ methods of properties.
 
-- Should methods be: to_bytes, as_bytes, tobytes, asbytes??
-
-- Plane.array_format could be a format string for array.array
-- Plane.update_from_array(array.array)
-    - If it were HD RGB then it would be 1920 * 1080 * 3 long.
-- Plane.update_from_ndarray(numpy.ndarray)
-    - If it were HD RGB then it would have shape (1080, 1920, 3)
-- Plane.update(input_)
-    And then it tries if it is a buffer, memoryview, bytes, etc..
-
-- TestCase.rms_diff(one, two) -> Root-mean-square diff
-- try to wrap API of testsrc filters
-
-- Replicate av_frame_get_best_effort_timestamp
+- [ ] TestCase.rms_diff(one, two) -> Root-mean-square diff
+- [ ] try to wrap API of testsrc filters
 
-    http://ffmpeg.org/pipermail/ffmpeg-devel/2011-February/104327.html
-    http://pastebin.com/Aq8eDZw3/
-    http://web.archiveorange.com/archive/v/yR2T4bybpYnYCUXmzAI5
+- [ ] Replicate av_frame_get_best_effort_timestamp
+    - http://ffmpeg.org/pipermail/ffmpeg-devel/2011-February/104327.html
+    - http://pastebin.com/Aq8eDZw3/
+    - http://web.archiveorange.com/archive/v/yR2T4bybpYnYCUXmzAI5
 
 
 ```
\ No newline at end of file